Department of Genetics, SK-50, University of Washington, Seattle, Washington 98195. Manuscript received July 22, 1983. Revised copy accepted September 24 ...
Copyright 0 1984 by the Genetics Society of America
ENHANCED GENE CONVERSION AND POSTMEIOTIC SEGREGATION IN PACHYTENE-ARRESTED SACCHAROMYCES CEREVISIAE LANCE S. DAVIDOW'
AND
BRECK BYERS
Department of Genetics, SK-50, University of Washington, Seattle, Washington 98195
Manuscript received July 22, 1983 Revised copy accepted September 24, 1983 ABSTRACT
Previous study has demonstrated that incubation of yeast cells of strain AP1 in sporulation medium at 36" permits them to begin meiosis but that they become arrested at pachytene and undergo enhanced intragenic recombination between ade2 heteroalleles. Tetrad analysis was undertaken to characterize the altered program of meiotic recombination more widely. In one set of experiments, pachytene-arrested cells were permitted to resume sporulation upon transfer to the permissive temperature. In the resulting asci, both postmeiotic segregation and gene conversion were increased several-fold at a number of loci relative to unarrested controls, whereas reciprocal recombination increased two- to threefold. Another set of experiments analyzed the genetic consequences of inducing the pachytene-arrested cells to revert directly to mitotic growth without completion of meiosis. The appearance of homozygous sectors from heterozygous markers revealed that these cells had become committed to appreciable recombination but that reciprocal exchange was less frequent than in normal asci. Taken together, the data indicate that pachytene arrest rendered the cells committed to enhanced recombination upon resumption of sporulation but that most of the crossing over did not occur until release from the arrest. -The genetic basis of pachytene arrest by AP-I was investigated by mating each of its parents with progeny of strain Y55, which is able to sporulate at 36". Both of these diploids sporulated at 36", and asci from the one studied further exhibited 2:2 segregation of the sporulation defect, indicating that pachytene arrest is dependent on a recessive, temperature-sensitive allele at a chromosomal locus.
LTHOUGH many of the mechanisms essential to meiotic recombination A remain obscure, we have gained considerable understanding by integrating findings from a variety of organisms. Data from the Ascomycetes have proven particularly useful because all of the meiotic products are recovered together. Tetrad analysis has demonstrated that crossing over is often accompanied by nonreciprocal recombination, or gene conversion (scored as 1:3 and 3:1, or 2:6 and 6:2 asci), and postmeiotic segregation (usually scored as 3:5 and 5:3 asci). These findings stimulated the concept that recombination entails the formation of a Holliday intermediate, or crossed strand-exchange form, in which single strands of DNA become base paired with homologous partners on either side of a common junction (HOLLIDAY1964). Electron microscopy
' Present address: Pfizer Central Research, Genetics 106: 165-183 February
1984.
Croton, Connecticut
06340.
166
L.
S. DAVIDOW AND
B.
BYERS
has revealed the existence of such forms in recombining bacteriophage DNA (VALENZUELA and INMAN1975) and in some eukaryotic plasmids, including the 2-pm DNA plasmid in meiotic yeast (BELLand BYERS1979). Further insight has been gained from extensive tetrad analysis in Saccharomyces cerarisiae. Here, gene conversion and postmeiotic segregation events are accompanied in approximately half of the tetrads by the reciprocal recombination of flanking markers (FOCELet al. 1979), confirming the association of hybrid DNA with crossing over. These tetrad data also indicate that hybrid molecules are not formed in reciprocal pairs, such as would result from branch migration after formation of the crossed strand-exchange form. It may, therefore, be the case that most hybrid DNA is formed by invasion of one duplex by a single strand displaced from the other (MESELSONand RADDINC1975). Nevertheless, alternative models have not been excluded and the molecular mechanisms of crossing over remain obscure (STAHL1979; SZOSTAK et al. 1983). We have gained a better understanding of meiotic recombination from the analysis of meiotic mutants, which have been isolated in a variety of organisms (BAKERet al. 1976). Studies of female Drosophila have, for example, revealed that mutants defective in crossing over can be identified by the nondisjunction of nonexchange bivalents in the first meiotic division (LINDSLEY and SANDLER 1977). Many of these mutants are also defective in the distribution of recombination events, as might result from failure at an early stage of crossing over. In contrast, mei9 maintains a normal distribution despite a marked decrease in reciprocal recombination. CARPENTER (1982) has recently found that this mutation also causes a substantial increase in intragenic recombination, presumably because an abnormally large amount of hybrid DNA is formed. It is an intriguing possibility that the enhancement of intragenic recombination results from excess activity by the same mechanism that results in gene conversion at sites of crossing over in the fungi. Mutational analysis of meiosis has also been undertaken in yeast and has resulted in the characterization of several spo mutations, which fail at various phases of sporulation (ESPOSITOand ESPOSITO1978), as well as rec (RODARTERAMON1972), mei (ROTH 1973) and con (FOCELand ROTH 1974) mutants, which may be defective more specifically in recombination processes. Certain radiation-sensitive (rad) mutations also have specific defects in meiotic recombination (GAMEet al. 1980). In addition, many of the cdc mutants are unable to complete sporulation for a variety of reasons (SIMCHEN1974). Analysis of the recombinational effects of these and other mutants has employed the capability of yeast to undergo reversion to mitosis after having become committed to meiotic recombination (SHERMAN and ROMAN1963). Using this assay, ZAMB and ROTH (1977) investigated the effect of blocking DNA synthesis by exposing a cdc8 homozygote to 36" and found that meiotic gene conversion requires chromosomal DNA replication. SCHILDand BYERS(1978) demonstrated a similar failure of recombination in another replication-defective mutant, cdc21, which is defective in thymidylate synthetase; on the other hand, cdc7, which is defective in the initiation of DNA replication in mitosis, prevented gene conversion without inhibiting premeiotic DNA replication.
ENHANCED CONVERSION AND PMS IN YEAST
167
In the course of these studies, we found that cells arrested by various cdc mutations, although capable of reverting to mitosis, are unable to resume meiosis and form complete asci of viable spores when transferred to the permissive temperature. On the other hand, a "wild-type" strain (AP-I) became arrested in pachytene upon transfer to 36" but remained capable (when returned to 23") either of reverting to mitosis or resuming sporulation (BYERS and GOETSCH1982). Assays for the production of adenine prototrophs by intragenic recombination between ade2 heteroalleles revealed enhanced gene conversion among cells recovering from the arrest by either route. Because of the possibility that an understanding of this phenomenon might shed light on the recombinational mechanism, we have sought to examine the genetic consequences of the arrest in more detail. The tetrad data reported here demonstrate that pachytene arrest is accompanied by widespread alterations of both reciprocal and nonreciprocal recombination. MATERIALS AND METHODS Strains: Strain AP-1 is the product of a cross between A364a (MATa adel ade2-1 u r d tyrl his7 lys2 g a l l ) and (~131-20(MAT@ ade2-R8 cyh2 can1 leu1 u r d MALx) (HOPPERet al. 1974). Strain Y55 (GARVIK and HABER1978) was provided by J. HABER. Media: YEPD medium contained 20 g/liter of glucose, 20 g/liter of bactopeptone and 10 g/ liter of yeast extract. YPA presporulation growth medium contained 10 g/liter of potassium acetate, 20 g/liter of bactopeptone and 10 g/liter of yeast extract. SPM sporulation medium contained 3 g/liter of potassium acetate and 0.2 g/liter of raffinose. Diagnostic media, sporulation medium for plates and the standard methods for testing genetic markers have been described by MORTIMER and HAWTHORNE (1 975). Sporulation: Cultures were grown in YPA by rotary agitation in flasks at 23" to a density of approximately 8 X lo6 cells/ml (monitored by optical density). They were harvested by centrifugation, washed in water, resuspended in SPM at the original cell density and again incubated on the rotary shaker. Reversible pachytene arrest at 36":T o arrest AP-I cells in pachytene, cultures were placed at 36" upon transfer to SPM. (Although the restrictive temperature will be designated as 36" in this report, a temperature of 36.5" was often employed.) Progress of the cells through meiosis was assayed by electron microscopic observation of cell lysates spread on an aqueous surface (GOETSCH and BYERS1982). Pachytene was defined in these spreads as the presence of synaptonemal complex from telomere to telomere in all chromosomes. The efficiency of sporulation was determined by phase-contrast microscopy, examining 200 cells at 500X for each determination. A finding of -no sporulation" signifies that no asci were seen in approximately 1000 cells. Shifting the arrested cells from 36" to 23" permitted reversal of the arrest and resumption of sporulation. We have previously shown (DAVIWW,GOETSCHand BYERS1980) that prolonged arrest of meiosis at 36" before reversal resulted in a decrease in the average number of spores per ascus, whereas supplementation of the arrested cells with fresh SPM increased the proportion of four-spored asci without significantly changing the frequency of recombinant adenine prototrophs or postmeiotic segregants. Therefore, we added fresh SPM at the time of the shift for the largest ascus dissection experiment presented in RESULTS. Transfer of the arrested cells to YEPD permitted resumption of mitotic growth. Treatment with ultrasound at 100 watts for 10 sec did not significantly affect viability but enhanced the separation of cells for micromanipulation and platings. Spore treatments: For random spore preparations, the SPM cultures were first washed in water and pretreated for 10 min in 0.1 M 2-mercaptoethanol, 0.02 M ethylenediaminetetraacetic acid and 0.2 M tris(hydroxymethy1)aminomethane hydrochloride, pH 9. They were then incubated in glusulase (Endo Laboratories) at a 1/10 dilution in water for 1 hr to remove ascal walls and lyse vegetative cells. Reconstruction experiments demonstrated that fewer than one in 1O6 vegetative
168
L. S. DAVIDOW AND B. BYERS
cells survived this treatment. The spores were sonicated on ice 3 X 15 sec at 100 watts before dilution to disperse clumps and were plated on YEPD or adeninedeficient growth media to measure the prototroph frequency. The method of DAWESand HARDIE(1974) for ether-killing vegetative cells was used to detect segregation of the ability to sporulate at 36" in some crosses. Sporulation plates were replica plated to YEPD agar in glass Petri dishes, which were then exposed to ether vapor in a closed vessel for 3-5 hr. Only sporulated colonies survived the treatment and grew into dense patches. The individual ascus method of dissection was described previously (DAVIDOW,GOETSCHand BYERS1980). Sporulated cultures were pretreated (as described before) and streaked on YEPD dissection plates. Individual, intact, four-spored asci were micromanipulated onto a segment of the plate coated with 1/5 X glusulase to digest the ascal walls. Each ascus was then dissected on the same plate. Dissection plates were used directly as master plates in replica plating to score marker segregation (FOGELet al. 1979). Scoring for ural, ura3, adel, gall and MAT required prior crossstamping with appropriate testers (two pairs of strains contained the needed combination of markers), whereas the other markers were scored directly on the first replica. The detection of postmeiotic segregations (PMS) as sectored colonies may have been less efficient for the markers requiring cross-stamping. RESULTS
The 36" pachytene arrest of AP-1 cells in sporulation medium could be reversed by either of two pathways. Resumption of meiotic sporulation was induced by shifting the culture to a lower (permissive) temperature while leaving the cells in SPM. Alternatively, transfer of the cells to YEPD growth medium at either temperature permitted return to vegetative growth without completion of sporulation. The recombinational effects of reversible pachytene arrest were assayed after both pathways of recovery. Effect of the pachytene arrest on recombination in asci Four aspects of genetic recombination were examined in the asci from reversibly arrested cells: (1) heteroallelic recombination between two mutant alleles to generate a wild-type gene (at ade2), (2)gene conversion (non-2:2 ratios) at 12 heterozygous loci assayed in dissected tetrads, (3) reciprocal recombination between linked genes and in gene-centromere intervals, and (4)PMSthe generation of half-sectored colonies from single ascospores. Heteroallelic recombination: The frequency of adenine prototrophic spores from reversibly arrested cells increased with time of exposure to the restrictive temperature (36")up to approximately eight or nine times the level in control spores not arrested at pachytene (Table 1). A cytological time c3urse (Table 2) revealed that significant numbers of pachytene figures were first present in these cultures at 9 hr after transfer to SPM at 36"; by 12 hr virtually all of the cells that were undergoing meiosis had reached pachytene. We conclude from these observations that cells subjected to reversible arrest at 36" undergo enhanced levels of intragenic recombination only if they are left at the elevated temperature until after they have entered pachytene. Gene conuersion and PMS in dissected tetrads: More than 100 asci with four viable spores were dissected and analyzed from each of three parallel subcultures exposed to the restrictive temperature for 0, 9 or 21 hr. To eliminate the possibility of false tetrads, preselected asci were individually micromanipulated to selected positions on the dissection plate before wall removal and
169
ENHANCED CONVERSION AND PMS IN YEAST TABLE I Heteroallelir recoinbi,zation in spores froin reversibly arrested cells Experiment I
Experiment I1
Time at 36" (hr)
Adenine prototrophs (per IO6 viable spores)
0 9 15 21
39 44 260 290
Time at 36" (hr)
Adenine prototrophs (per 1 O6 viable spores)
0 7 9 11 13 15 17 19 21
42 45 55 144 145 244 303 296 397
Cultures of AP-1 were placed in SPM at 36" for the indicated time before shifting to 23" to complete sporulation. Cultures remained in SPM a total of 48-72 hr before harvest. TABLE 2 Cytological time course at the restrictive temperature ~
~
~~~~
Time at 36" (hr)
% nuclei in pachytene
0 3 6 9 12
0 0 0 12 76
Time points for electron microscopic cytology were taken from an AP-1 culture transferred to SPM at 36" at T = 0. The first 50 nuclei observed were scored from each time point.
ascospore dissection (see MATERIALS AND METHODS). The segregations of each of 12 markers in the tetrads (Table 3) showed a dramatic increase in non-2:2 segregation at all loci in the reversibly arrested preparation. In asci from cells left at 23" or downshifted after only 9 hr at 36", all markers showed a low level of non-2:2 segregation, ranging from 0 for several loci to 2% for leu1 and canl. All of the markers showed increased non-2:2 segregation in the culture from 21 hr at 36" with frequencies ranging from 4% for gall and MALx to 21% for u r a l . The increase in non-2:2 segregation averaged over all of the markers was approximately 15-fold. PMS in spores from reversibly arrested pachytene cultures also showed a particularly striking increase over the controls (Table 4). T h e distribution of these postmeiotic segregations among the different heterozygous markers was nonuniform. More than half of the 126 PMS occurred at lys2 or his7, whereas five of the markers, including mating type, had none. Overall, PMS increased about 20-fold in the 21-hr group; 41 PMS occurred in 404 spores from asci with all four spores viable, whereas the incomplete asci had a slightly, but not significantly, higher rate of 85 PMS in 726 viable spores.
170
L. S. DAVIDOW AND B. BYERS
TABLE 3 Not1-2:2 segregotiotts
it1
asci from reversibly arrested cells
A. Cultures with less than 10 hr at 36" (234 asci) 9 hr at 36" (123 asci) Marker
Always at 23" (1 11 asci)
3: 1
1:3
3: 1
5:3
1:3
0 0 0 0 0 0 0 1
1 0 0
0 0 0 0 0 0 0
1 0
1 0 2 0 8
1 0 0 0 0 0 0 1 0 0 0 0 2
1 0 0 0 0 0 1 0 2 0
ndel gall lys2 tyr1 his7 MAT" ura? cu )Z 1
leu 1 ryh2 ural MAL Totals
1 0 0 1
2
0 0 2
1 1 0
0 0 2
Combined % non-2:2
1.3 C0.4 C0.4 0.4
